MENU

Click here to

Ă—

Are you sure ?

Yes, do it No, cancel

Gray Matter-Based Radiomics and Machine Learning for the Diagnosis of Attention-Ddeficit/Hyperactivity Disorder

S Zhao1*, Z Mu2, H Zhao3, J Qiu3, W Lu3, W Lu3, L Shi3, (1) Beijing Anding Hospital, Capital Medical University, Beijing, CN, (2) The Second Affiliated Hospital Of Shandong First Medical University, Taian, CN, (3) Shandong First Medical University & Shandong Academy Of Medical Sciences, Taian, CN

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: community has no objective biological tools capable of diagnosing attention-deficit/hyperactivity disorder (ADHD). This study aimed to investigate whether the gray matter-based radiomics and machine learning can be used to diagnose ADHD.


Methods: study collected 155 participants’ structural MRI data of New York University (NYU) Medical Center from the ADHD-200 Global Competition, including 67 typically developing children (TDC) and 88 patients with ADHD. We performed image preprocessing, including image reconstruction, correction, registration and segmentation, using cat12 software based on SPM12 in Matlab 2013. The standardized gray matter volume images were exported after the preprocessing was finished by cat12. A total of 1057 radiomics features were extracted from the whole gray matter using IBEX source, including first-order statistical features and high-order texture features calculated based on gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM) and neighborhood intensity difference matrix (NIDM). The two-sided Wilcoxon rank sum test was used to calculate the differences between TDC group and ADHD group; the features with P < 0.05 were selected for further analysis. We further selected the predictive features and classified the two groups using a sequential backward elimination support vector machine (SBE-SVM) algorithm. The 155 participants were divided into the training set (n = 117), the validation set (n = 13) and the test set (n = 25) in this procedure.


Results: gray matter-based radiomics model was able to discriminate those with ADHD from the TDC (accuracy: 72%; area under curve: 0.85; sensitivity: 80%; specificity: 60%). This model finally selected 24 predictive radiomics features, including 20 GLCM-features and 4 first-order features (P = 7x10?5-0.049).


Conclusion: study demonstrates that the gray matter-based radiomics machine learning model can provide discrimination of those with ADHD from TDC. The gray matter-based radiomics may be the potential biomarker for the clinical diagnosis of ADHD.

Funding Support, Disclosures, and Conflict of Interest: This study was supported by the Shandong Province Key Research and Development Program (2017GSF218075) and Taishan Scholars Program of Shandong Province.

Keywords

MRI, Brain, Feature Extraction

Taxonomy

IM- MRI : Machine learning, computer vision

Contact Email